This thesis is centered upon an optimal trajectory generation algorithm that allows real-time control for cooperation of multiple quadrotor vehicles for intelligence, surveillance, and reconnaissance missions with minimal user input. The algorithm is designed for an indoor environment where global positioning system data is unavailable or unreliable, forcing the vehicles to obtain position data using other sensors. This thesis specifies the lab setup and well as the control approach used. Data acquired from two experiments is included to demonstrate the effectiveness of the control approach. The control approach described within allows for a fully autonomous system with user input required only at the initiation of a mission. The algorithm blends trajectory planning, trajectory following, and multi-vehicle coordination to achieve the goal of autonomy. The focus of the thesis was on trajectory generation and multi-vehicle coordination, while leveraging existing trajectory following controller implementations. The trajectory generation is accomplished with a direct transcription of the optimization problem that leverages inverse dynamics and separates spatial and temporal planning. The vehicle motion is constrained, and simplifying multi-vehicle coordination assumptions allow for the efficient solution and execution of the problem.